Exploring the long-term impact of a nurse-midwife mentorship intervention in Neno district, Malawi: a secondary data analysis of maternal and neonatal complications.

Background: There is critical need to strengthen the global nursing and midwifery workforce. This is especially true in Malawi where they are the primary providers of obstetric and neonatal care. In Neno district, Malawi, in 2017, we implemented an intensive training and longitudinal bedside mentorship intervention for nurses and midwives. From inception, there was a pre-planned project target completion after 5 years, including a staged handover to local ownership after 3 years. The objective of this study was to assess differences in reported maternal and neonatal complications following project completion and handover to local leadership.

Methods: The project was a partnership between an academic institution and local nongovernmental organization. In October 2020, the intervention was handed over and maintained by the local organization with limited support from the academic institution. Data from January 2019 to May 2023 were extracted from the Malawi District Health Information Software 2. Bivariate analyses explored differences in the pre- and post-handover periods for all government-collected, birth-related variables. The "pre-handover" period encompassed January 2019 to September 2020 and "post-handover" from October 2020 to May 2023. Multivariate linear regression explored outcomes while controlling for health facility.

Results: Data were collected from 10 health facilities in the district and included a total of 23,259 births. Overall, there were few significant changes between periods. Exceptions were in reporting of antepartum hemorrhage (p < 0.01), maternal sepsis (p = 0.01), and manual removal of the placenta (p < 0.01), where we observed decreases in reporting. There was a significant decrease in the reporting of neonatal sepsis (p = 0.01) in the bivariate analysis, which remained only borderline significant in the multivariate model (p = 0.05). Where differences occurred, they were associated with improvements in reported identification of complications and provision of associated emergency care.

Conclusions: Few differences in reported maternal and neonatal complications between the periods suggest positive impact of the intervention was sustained following project handover and transition of activities to local leadership. These findings strengthen support for longitudinal mentorship as a pivotal component for skill retention after training. Transparent partnerships which include pre-determined end points and time for handover of activities to local ownership are crucial components for sustainability.